End of training

README.md

@@ -2,7 +2,10 @@
 license: apache-2.0
 base_model: bert-large-uncased
 tags:
+- question-answering
 - generated_from_trainer
+datasets:
+- squad
 model-index:
 - name: out
   results: []
@@ -13,7 +16,7 @@ should probably proofread and complete it, then remove this comment. -->
 
 # out
 
-This model is a fine-tuned version of [bert-large-uncased](https://huggingface.co/bert-large-uncased) on the None dataset.
+This model is a fine-tuned version of [bert-large-uncased](https://huggingface.co/bert-large-uncased) on the squad dataset.
 
 ## Model description
 

all_results.json

@@ -1,11 +1,11 @@
 {
-    "epoch": 20.0,
+    "epoch": 50.0,
     "eval_samples": 3000,
-    "exact_match": 67.2,
-    "f1": 77.09945992044638,
-    "train_loss": 0.40876169204711915,
-    "train_runtime": 950.4852,
+    "exact_match": 68.3,
+    "f1": 77.35539073416115,
+    "train_loss": 0.1694981136669715,
+    "train_runtime": 2384.026,
     "train_samples": 3000,
-    "train_samples_per_second": 63.126,
-    "train_steps_per_second": 0.505
+    "train_samples_per_second": 62.919,
+    "train_steps_per_second": 0.503
 }

args.bin

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:f471e6f312bda43d0ba2e1d68076024927158d178f3e43abe3b9098c86051796
+oid sha256:4690bba900b977f370315e9fca36d8a47a29d3863a90bbcf8c577c9fd20ac63a
 size 5652

eval_nbest_predictions.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:0149ec0ca7bfeee41220a704e7e726b3648029a0c4fdd1dbc13d44e48846b7f0
-size 13780699
+oid sha256:6763bbcbd8b21d7e3fed94b03bacdb4881ee275f634f08a710c42512c94e46f3
+size 13811032

eval_results.json

@@ -1,6 +1,6 @@
 {
-    "epoch": 20.0,
+    "epoch": 50.0,
     "eval_samples": 3000,
-    "exact_match": 67.2,
-    "f1": 77.09945992044638
+    "exact_match": 68.3,
+    "f1": 77.35539073416115
 }

train_results.json

@@ -1,8 +1,8 @@
 {
-    "epoch": 20.0,
-    "train_loss": 0.40876169204711915,
-    "train_runtime": 950.4852,
+    "epoch": 50.0,
+    "train_loss": 0.1694981136669715,
+    "train_runtime": 2384.026,
     "train_samples": 3000,
-    "train_samples_per_second": 63.126,
-    "train_steps_per_second": 0.505
+    "train_samples_per_second": 62.919,
+    "train_steps_per_second": 0.503
 }

trainer_state.json

@@ -1,28 +1,40 @@
 {
   "best_metric": null,
   "best_model_checkpoint": null,
-  "epoch": 20.0,
+  "epoch": 50.0,
   "eval_steps": 500,
-  "global_step": 480,
+  "global_step": 1200,
   "is_hyper_param_search": false,
   "is_local_process_zero": true,
   "is_world_process_zero": true,
   "log_history": [
     {
-      "epoch": 20.0,
-      "step": 480,
-      "total_flos": 4.179181464576e+16,
-      "train_loss": 0.40876169204711915,
-      "train_runtime": 950.4852,
-      "train_samples_per_second": 63.126,
-      "train_steps_per_second": 0.505
+      "epoch": 20.83,
+      "learning_rate": 2.916666666666667e-05,
+      "loss": 0.3975,
+      "step": 500
+    },
+    {
+      "epoch": 41.67,
+      "learning_rate": 8.333333333333334e-06,
+      "loss": 0.0083,
+      "step": 1000
+    },
+    {
+      "epoch": 50.0,
+      "step": 1200,
+      "total_flos": 1.044795366144e+17,
+      "train_loss": 0.1694981136669715,
+      "train_runtime": 2384.026,
+      "train_samples_per_second": 62.919,
+      "train_steps_per_second": 0.503
     }
   ],
   "logging_steps": 500,
-  "max_steps": 480,
-  "num_train_epochs": 20,
+  "max_steps": 1200,
+  "num_train_epochs": 50,
   "save_steps": 500,
-  "total_flos": 4.179181464576e+16,
+  "total_flos": 1.044795366144e+17,
   "trial_name": null,
   "trial_params": null
 }